ML-Based Transport Mode Detection for Population Distribution

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing systems face challenges in accurately determining transport modes and population distribution for buildings in a geographic region, which affects the accuracy of mobility patterns and related services such as navigation, infrastructure planning, and emergency responses.

Innovation Solution

A system utilizing a trained machine learning model processes sensor data from user equipment to determine transport modes and population distribution, updating map data and controlling geo-location services based on these determinations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If sensor data from mobile phones is collected and analyzed to determine transport modes and population distribution, then the accuracy of mobility patterns and related services is improved, but the complexity of the system increases

Engineering Contradiction:

Improveaccuracy of transport mode determinationVSAvoidsystem complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent introduces an intermediary processing system that aggregates sensor data from multiple mobile phones and processes it through machine learning models. This intermediary layer handles the complexity of data analysis, transforming raw sensor data into reliable transport mode classifications and population distribution metrics, thereby improving measurement precision while managing system complexity through centralized processing.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent replaces traditional mechanical or manual methods of determining transport modes and population distribution with automated machine learning-based systems. The system uses sensor data processing and algorithmic analysis to substitute complex manual surveying and counting methods, achieving higher accuracy while the system manages complexity through automated computational processes.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Reliability

If sensor data is collected from multiple user equipment to determine population distribution, then the reliability of population data is improved, but the loss of time for data processing increases

Engineering Contradiction:

Improvereliability of population distribution dataVSAvoiddata processing time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent implements preliminary action by continuously collecting and pre-processing sensor data in the background as users move through geographic regions. The system performs preliminary aggregation and filtering of data from multiple user equipment, preparing it for analysis before it is actually needed for population distribution determination. This reduces the time required for final processing while maintaining reliability through continuous data accumulation.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent applies partial action by selectively processing data from a subset of user equipment that provides sufficient statistical reliability. Rather than processing every single sensor data point from all users, the system identifies and processes representative samples that achieve the required reliability threshold, thereby reducing processing time while maintaining data quality.

Inventive Principle:

Principle #16Partial or excessive action

3Measurement precision

If machine learning models are used to determine transport modes and population distribution, then the accuracy of mobility patterns is improved, but the loss of information during data processing increases

Engineering Contradiction:

Improveaccuracy of mobility pattern determinationVSAvoidinformation loss in sensor data

Core Design Contradiction:

Measurement precisionVSLoss of information

Solution Approach 1:

The patent implements feedback mechanisms where the machine learning model's predictions are continuously refined based on comparison with actual observed patterns. The system uses feedback loops to adjust processing parameters and preserve critical information features that contribute to accurate transport mode classification and population distribution calculation, thereby maintaining measurement precision while minimizing information loss through iterative optimization.

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS11343636B2Automatic building detection and classification using elevator/escalator stairs modeling—smart cities

Publication Date: 2022.05.24 HERE GLOBAL BV

AI summary

A system, a method and a computer program product are provided to determine population distribution of users associated with one or more buildings in a geographic region, using a machine learning model. The system may include at least one memory configured to store computer executable instructions and at least one processor configured to execute the computer executable instructions to obtain mobility features associated with the one or more buildings in the geographic region. The processor may be configured to determine using a trained machine learning model, one or more transport modes for the one or more buildings, based on the mobility features. The processor may be further configured to determine, using the trained machine learning model, the population distribution of the users associated with the one or more buildings in the geographic region at a fixed epoch based on the determined one or more transport modes.